Successfully reported this slideshow.
We use your LinkedIn profile and activity data to personalize ads and to show you more relevant ads. You can change your ad preferences anytime.

Measuring polarization on social media

265 views

Published on

Joint work with Kiran Garimella, Gianmarco De Francisci Moralers, Aris Gionis.

Published in: Data & Analytics
  • My brother found Custom Writing Service ⇒ www.HelpWriting.net ⇐ and ordered a couple of works. Their customer service is outstanding, never left a query unanswered.
       Reply 
    Are you sure you want to  Yes  No
    Your message goes here
  • You can hardly find a student who enjoys writing a college papers. Among all the other tasks they get assigned in college, writing essays is one of the most difficult assignments. Fortunately for students, there are many offers nowadays which help to make this process easier. The best service which can help you is ⇒ www.WritePaper.info ⇐
       Reply 
    Are you sure you want to  Yes  No
    Your message goes here
  • If u need a hand in making your writing assignments - visit ⇒ www.HelpWriting.net ⇐ for mire detailed information.
       Reply 
    Are you sure you want to  Yes  No
    Your message goes here
  • High paying Twitter jobs? $25 per hour, start immediately ★★★ http://t.cn/AieX6y8B
       Reply 
    Are you sure you want to  Yes  No
    Your message goes here
  • Be the first to like this

Measuring polarization on social media

  1. 1. measuring polarization on social media michael mathioudakis computer science aalto university http://www.michalis.co
  2. 2. 2 social media Michael Mathioudakis consume content news about friends, politics, favorite artists generate content share experiences, interesting articles interact with others comment, rate, and discuss hundreds of millions of active users
  3. 3. why study social media a window into the thoughts and actions of people how people spend their time what attracts their attention what they think about political issues how they interact socially unique opportunity for social scientists study human behavior at large scale & fine detail Michael Mathioudakis 3
  4. 4. but you are a computer scientist develop algorithms to extract insights from social media two challenges 1- volume of data 2- complexity of social behavior calls for complex models (e.g., graphs) that require fast algorithms solution algorithmic methods, software tools and systems for large-scale computation analyze efficiently & automatically large amounts of data Michael Mathioudakis 4
  5. 5. 5 no information barriers citizen journalism social connectivity … … harassment fake news polarization social media: good and bad sides Michael Mathioudakis
  6. 6. polarization political or social polarization ‘the act of separating or making people separate into two groups with completely opposite opinions’* related term: controversy ‘public discussion and argument about something that many people strongly disagree about’* *oxford english dictionary Michael Mathioudakis 6
  7. 7. why study polarization? polarization can be linked to adverse effects social segmentation stereotypes echo chambers goal understand and mitigate them Michael Mathioudakis 7
  8. 8. 8 0 10 20 30 40 50 60 70 80 90 100 Donald Trump Approval Rating % February 2017 Democrats Republicans Michael Mathioudakis
  9. 9. 9 ideology of USA public source: Pew Research Center 19942014 Democrats and Republicans have drifted apart Michael Mathioudakis
  10. 10. why study polarization on social media extract insights on the polarization process some specific questions what are the polarizing issues? does polarization increase over time? we saw one such instance: US public how could we ‘nudge’ people away from extreme polarization? is polarization linked to echo chambers? do social media increase polarization? Michael Mathioudakis 10 this talk
  11. 11. in this talk… twitter algorithms to measure it long-term study next steps 11 polarization Michael Mathioudakis
  12. 12. 12 retweets tweet replies connections interactions Michael Mathioudakis
  13. 13. twitter microblogging platform since 2006; 300 million active users users post short messages - ‘tweets’ 13Michael Mathioudakis
  14. 14. global structure of interactions nodes 14 user accounts edges retweets replies connections interaction graph Michael Mathioudakis 1% sample of all tweets from the internet archive data
  15. 15. in what follows… 15Michael Mathioudakis twitter algorithms to measure it long-term study next steps polarization Quantifying Controversy in Social Media. WSDM 2016. With Kiran Garimella, Gianmarco De Francisci Morales, Aristides Gionis.
  16. 16. how polarized is a discussion? e.g., obamacare in January 2017 16Michael Mathioudakis
  17. 17. how could we approach this question? content? do opposing sides say different things? sentiment? do polarized topics exhibit wider range of emotions? interactions? do people interact more with their own side? Michael Mathioudakis 17 let’s try this
  18. 18. method template build an interaction graph several types to try retweets, replies, connections is the interaction graph polarized? output: polarization score 18 not polarized polarized two sides, well separated Michael Mathioudakis
  19. 19. pipeline 19Michael Mathioudakis topic evaluation Graph Building Graph Partitioning Controversy Measure Polarization what type of interaction graph should we use? how to find two sides in the graph? how to measure the separation between two sides? do we identify polarized discussions?
  20. 20. retweets replies connections 20Michael Mathioudakis topic evaluation Graph Building Graph Partitioning Controversy Measure Polarization what type of interaction graph should we use? how to find two sides in the graph? how to measure the separation between two sides? do we identify polarized discussions?
  21. 21. 21 partition into two sides Michael Mathioudakis topic evaluation Graph Building Graph Partitioning Controversy Measure Polarization what type of interaction graph should we use? how to find two sides in the graph? how to measure the separation between two sides? do we identify polarized discussions?
  22. 22. partitioning many existing algorithms spectral clustering label propagation metis Michael Mathioudakis 22 well connected badly connected
  23. 23. random walk controversy (RWC) 23Michael Mathioudakis topic evaluation Graph Building Graph Partitioning Controversy Measure Polarization what type of interaction graph should we use? how to find two sides in the graph? how to measure the separation between two sides? do we identify polarized discussions?
  24. 24. random walk controversy (RWC) polarization score how difficult to reach c from B 24 A and B: two sides c and d: central nodes and values -1 to 1 the more difficult the higher the scorehow difficult to reach d from A Michael Mathioudakis c d A B random walk
  25. 25. random walk 25Michael Mathioudakis discrete process on a graph step 1, 2, 3, … at each step, we are at one node at next step, we move to a nearby node - following an edge at random specify where the walk starts (e.g., at c) where the walk ends (e.g., at d) c d
  26. 26. random walk controversy (RWC) 26 A and B: two sides c and d: central nodes Michael Mathioudakis c d A B random walk start: random node of A / B 50% chance that we do either end: at central nodes c of side A or d of side B PXY = P(started in X | ended in Y) consider two independent instances one ending at the central node of side A, the other at the central node of side B RWC = PAA PBB - PBA PAB both random walks started in the side where they ended both random walks started in a side other than where they ended values?
  27. 27. random walk controversy (RWC) 27Michael Mathioudakis topic evaluation Graph Building Graph Partitioning Controversy Measure Polarization what type of interaction graph should we use? how to find two sides in the graph? how to measure the separation between two sides? do we identify polarized discussions?
  28. 28. 28Michael Mathioudakis topic evaluation Graph Building Graph Partitioning Controversy Measure Polarization what type of interaction graph should we use? how to find two sides in the graph? how to measure the separation between two sides? do we identify polarized discussions?
  29. 29. evaluation non-polarized topics indian beefban, nemtsov protests, netanyahu US congress speech, baltimore riots, ukraine polarized topics germanwings plane crash, sxsw, mother’s day, jurassic world movie, national kissing day 29Michael Mathioudakis popular topics of 2015 graphs with 1 – 150 thousand nodes examples:
  30. 30. Graph Building Graph Partitioning Controversy Measure Polarization results Michael Mathioudakis 30 does the pipeline distinguish polarized from non-polarized topics? retweets replies connections two sides random walk controversy (RWC) topic (if we use retweets)yes!
  31. 31. results nemtsov protests indian beef ban sxsw conference germanwings plane crash interaction graphs: retweets 31Michael Mathioudakis polarized topics non-polarized topics high RWC low RWC
  32. 32. results 32 retweets replies interaction graphs for nemtsov protests Michael Mathioudakis
  33. 33. an algorithmic way to quantify polarization Michael Mathioudakis 33 retweets two sides random walk controversy (RWC) topic based on structure of interactions language-independent Graph Building Graph Partitioning Controversy Measure Polarization new method can be deployed in the wild
  34. 34. in what follows… 34Michael Mathioudakis twitter algorithms to measure it long-term study next steps polarization Graph Building Graph Partitioning Controversy Measure The Ebb and Flow of Controversial Debates on Social Media. ICWSM 2017. The Effect of Collective Attention on Controversial Debates on Social Media. WebScience 2017. With Kiran Garimella, Gianmarco De Francisci Morales, Aristides Gionis. Polarization
  35. 35. polarization over time has polarization around controversial topics increased over time? 35Michael Mathioudakis
  36. 36. polarization over time data 1% sample of all tweets from the internet archive September 2011 to September 2016 method for a given topic (e.g., obamacare) retrieve related tweets build an interaction graph for each day using retweets measure RWC score Michael Mathioudakis 36
  37. 37. RWC vs Time 37Michael Mathioudakis September 2011 September 2016
  38. 38. volume of activity spikes at major events 38Michael Mathioudakis
  39. 39. 39Michael Mathioudakis
  40. 40. does polarization spike with volume? measure rwc vs volume of activity 40Michael Mathioudakis
  41. 41. RWC vs Volume higher volume higher controversy 41Michael Mathioudakis higher volume higher volume
  42. 42. RWC vs Volume Michael Mathioudakis 42 higher volume higher controversy rwc
  43. 43. summary 43Michael Mathioudakis twitter algorithms to measure it long-term study next steps polarization
  44. 44. why study polarization on social media extract insights on the polarization process some specific questions what are the polarizing issues? does polarization increase over time? we saw one such instance: US public how could we ‘nudge’ people away from extreme polarization? is polarization linked to echo chambers? do social media increase polarization? Michael Mathioudakis 44 this talk next steps
  45. 45. thank you! 45Michael Mathioudakis

×